1 / 35

Analysis of BK(*) against Hadronic Breco

This analysis compares the performance of BK(*) decays against Hadronic Breco using BaBar Full Simulation and SuperB Fast Simulation. It explores the reconstruction efficiency and distributions of neutral and charged Breco modes.

Télécharger la présentation

Analysis of BK(*) against Hadronic Breco

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Analysis of BK(*) against Hadronic Breco Elisa Manoni INFN Sez. Perugia DGWG parallel session, Annecy general meeting March 17, 2010

  2. Outline • Hadronic Breco and February production • BaBar Full Simulation vs Fast Simulation in the BaBar configuration • Comparison between SuperB Detector geometry # 3 (DG_3) and SuperB Detector geometry # 4 (DG_4), w/o BWD EMC • Breco side • B K(*) signal side analysis

  3. Hadronic Breco reconstruction in FastSim • SemiExclusive reconstruction implemented in FastSim: PacHadRecoilUserPackage(based on BaBar BTauSemiExclUser code) • Breco side:limit the number of reconstructed modes according to their purity • Breco mode classification: neat : purity > 80% , neat≈O(10-4) clean : 50% < purity < 80% , clean≈O(10-3-10-2) dirty : 8%<purity<50% , dirty≈O(10-2)  reconstruct only neat+clean modes • Bsig side: • K+ • Ks(+-) • K*+(Ks+, K+0) • K*0(K+-)  • + , with +e+, +, +, +(+0), a1+(0+) discussed in this talk

  4. February production in this talk: - signal MC samples w bkg - generic MC samples w/o bkg • two SuperB Detector Geometryused (seen next slides) +BaBar geometry (and beams) to compare Fast Sim with BaBar Full Sim • Generic samples (B+B-_generics, B0B0bar_generics, uds, ccbar) • DG_BaBar: w background, 50million evts • DG_3: w background (25M evts) + w/o background on (530M evts) • D_4: w background on (28million evts) + background on (830M evts) • Signal Samples:

  5. Detector geometries (I) +fwd fwd CsI + LYSO+ bwd 5 • DetectorConfiguraztion_3 • SVT_L0 + fwd DCH+ bwd EMC

  6. Detector geometries (II) +fwd 6 • DetectorConfiguraztion_4 • SVT_L0 + fwd PID + bwd EMC

  7. Analysis strategy • Compare DG_3 vs DG_4 • PID in the fwd region weaker in DG_3 (dE/dx) then in DG_4 (dE/dx+TOF) • in DG_3, smaller amount of material in front of fwd EMC • Use bwd EMC as VETO device: • reject candidates with neutrals reconstructed in bwd EMC • apply BaBar selection • require “zero” EExtraNeutralBwd • pid device angular acceptance: 20°(tracking)<qlab< 144.4 (DIRC) • Apply BaBar Analysis cuts for B+K+nn, B+K*+nn, B0K*0nn and compare efficiencies for Breco and Bsig selection

  8. A remark on PID usage 8 • PID selectors used in the D and B lists • p from GoodTracksVeryLoose • K PID: * Breco side, DIRC region: D KY lists use GoodTracksLoose or KLHNotPion depending on the D mode, B D(*)KY lists use KNNTight depending on the B mode * Breco side, Fwd region: KLHNotPion * Bsig side: KNNTight • select events in which the K pass the proper PID selector • kaons in the DIRC region: TableBased selectors (performances form BaBar) • kaons in the fwd region: * TableBased selectors with performances from TOF FullSim studies (DG_4) * TableBased selectors with performances from BaBar (DG_BaBar, DG_3)

  9. BaBaf Full Simulationvs SuperB Fast Simulation

  10. Fast Sim DG_BaBar vs BaBar Full Sim (I) • SuperB FastSim: • B+B-, B0B0bar, ccbar, uds MC samples • B+K+nn, B+K*+nn, B0K*0nn signal MC samples • BaBar beams and detector geometry • BaBar FullSim, Run3 (same code and same “skim” as in FastSim): • B+B- : 49,766x103 gen. evts • B0B0bar : 50,556x103 gen. evts • ccbar : 83,974x103 gen. evts • uds : 66,892x103 gen evts • B+K+nn/B+K*+nn/B0K*0nn: 7,845/7,8510/6,282 x 103 gen evts • Selection applied: • at least one reconstructed Breco; if #Breco > 1,best candidate  |DE|min • -0.09<∆E<0.05 GeV • 5.270<mES<5.288 GeV/c2

  11. Breco efficiencies, generic samples (I) • = nsel/nbreco(purity >0.5, abs(charge)==0/1, pid requirements) • (see back up for stat errors on efficiencies) loosing efficiency in the “signal sample”: B0B0bar for neutral Breco and B+B- for charged Breco

  12. neutral Breco shapes, generic samples • mES and DE before the selection * FullSim __ FastSim

  13. charged Breco shapes, generic samples • mES and DE before the selection * FullSim __ FastSim

  14. Bsig efficiencies, signal samples B0K*0 B+K*+(Ksp) B+K*+(Kp0)

  15. SuperB Fast Simulation: DG_3 vs DG_4

  16. neutral Breco efficiencies, generic samples • = nsel/ngen • (see back up for stat errors on efficiencies)

  17. charged Breco efficiencies, generic samples • = nsel/ngen • (see back up for stat errors on efficiencies)

  18. neutral Breco distributions, generic samples DG_BaBar DG_3 DG_4

  19. charged Breco distributions, generic samples (II) DG_BaBar DG_3 DG_4

  20. B+K+: selection efficiency (I) • BaBar-like cut and count analysis Qtag = ± 1 5.270<mES<5.288 GeV/c2 |cosBreco,Thrust|<0.85 K candidate from Bsig |cos*trk|<0.85 NextraTrk<3 Eextra<0.4 GeV N0=0 pKB > 1.1Gev/c -0.85<cospmiss<0.9 in the following,  = n_selected/n_generated

  21. B+K+: selection efficiency(II) eDG_3 / eDG_4 |skim= 1.03 eDG_3 / eDG_4 |costh_K= 1.04 eDG_3 / eDG_BaBar |skim= 2.21 eDG_3 / eDG_BaBar |costh_K= 1.24 extra p0 =0 Eextra<0.4 GeV

  22. B+K+: some distributions

  23. BK*: selection efficiency (I) • BaBar-like cut and count analysis Bsig-Breco charge correlation 5.270<mES<5.288 GeV/c2 -0.09<∆E<0.05 GeV |cosBreco,Thrust|<0.9 < 0.85 • in the following, • e= n_selected/n_generated • n.b.: very small stat, e (after K*mass cut) • compared in next pages consistent within stat errors E*miss+cp*miss>4.5GeV Eextra<0.3GeV

  24. B0K*0(Kp): selection efficiency (II) eDG_3 / eDG_4 |skim= 1.00 eDG_3 / eDG_4 |K*mass= 0.09 eDG_3 / eDG_BaBar |skim= 2.13 eDG_3 / eDG_BaBar |K*mass= 0.98 Eextra<0.3 GeV

  25. B+K*+(Kp0): selection efficiency (II) eDG_3 / eDG_4 |skim= 1.01 eDG_3 / eDG_4 |costh_K= 0.90 eDG_3 / eDG_BaBar |skim= 2.23 eDG_3 / eDG_BaBar |costh_K= 1.00 Eextra<0.3 GeV

  26. B+K*+(Ksp): selection efficiency (II) eDG_3 / eDG_4 |skim= 1.01 eDG_3 / eDG_4 |costh_K= 0.84 eDG_3 / eDG_BaBar |skim= 2.23 eDG_3 / eDG_BaBar |costh_K= 1.09 Eextra<0.3 GeV

  27. BK*: some distributions B+K*+ signal MC B0K*0 signal MC B0K*0 signal MC B+K*+ signal MC

  28. Remarks on Eextra shape Fast Sim B+K+ signal MC • background (radiative BhaBha) dramatically increase the number of reconstructed neutrals  Eextra shifts at high values, loosing the bin 0 discriminating power (apply a cut on the maximum gamma energy?) • in this production, not enough generic statistics with machine in, to study the bkg Eextra shape (probably shifted  need to enlarge the signal region to be more discriminant) • not enough signal statistics to quantify the benefits from the Bwd EMC veto • bkg studies to find minimum photon energy for the bwd • optimize Eextra_bwd cut Full Sim B+K+(Ksp) signal MC generics

  29. Conclusions • PacHadRecoilUser used in February production, in this talk • generic Mc samples w/o background • signal MC samples : BK+, BK*0, BK*+ with background • comparison with BaBar FullSim: • agreement in shapes and efficiency is improving (more details at the fast sim session) • test SuperB detector geometry configuration • DG_4 and DG_3 seems to be equivalent • adding radiative bhabha, the Eextra shape in signal-MC is less signal-like • need a high statistic generic sample, mixed with bkg, to evaluate generic Eextra shape • some attempts to improve this (Emax,g?), issue discussed also within the EMC group • not enough statistics to test the impact of bwd EMC

  30. Back-up slides

  31. Hadronic Breco reconstruction philosophy 31 • Aim: collect as many as possible fully reconstructed B mesons in order to study the property of the recoil • SemiExclusive reconstruction: search for B  D(*)X, with X=np mK pKs qp0 and n+m+r+q<6,without making requirements on intermediate resonances • Reoconstruction steps: • reconstruct D hadrons • use D as a seed and add X to have a system compatible with the B hypotesys • Signal box defined by using: • Sample of 1100 B decay modes, ordered by purity. • In events with multiple candidates, the best one is selected according to the smallest DE

  32. neutral Breco efficiencies, generic samples *** fast sim: B0B0Bar : mES cut = 0.2095+/-0.0010 de cut = 0.1840+/-0.0010 B+B- : mES cut = 0.1110+/-0.0011 de cut = 0.0909+/-0.0010 cc : mES cut = 0.1013+/-0.0012 de cut = 0.0812+/-0.0011 uds : mES cut = 0.1213+/-0.0021 de cut = 0.0879+/-0.0006 *** full sim: B0B0Bar : mES cut = 0.2548+/-0.0009 de cut = 0.2279+/-0.0008 B+B- : mES cut = 0.1156+/-0.0009 de cut = 0.0929+/-0.0008 cc : mES cut = 0.0967+/-0.0006 de cut = 0.0740+/-0.0005 uds : mES cut = 0.1246+/-0.0015 de cut = 0.0877+/-0.0004

  33. charged Breco efficiencies, generic samples *** fast sim: B0B0Bar : mES cut = 0.1403+/-0.0013 de cut = 0.1096+/-0.0012 B+B- : mES cut = 0.2895+/-0.0012 de cut = 0.2416+/-0.0012 cc : mES cut = 0.1284+/-0.0011 de cut = 0.0902+/-0.0009 uds : mES cut = 0.1370+/-0.0011 de cut = 0.09387+/-0.0003 *** full sim: B0B0Bar : mES cut = 0.15187+/-0.0010 de cut = 0.1183+/-0.0009 B+B- : mES cut = 0.3557+/-0.0003 de cut = 0.3094+/-0.0003 cc : mES cut = 0.1263+/-0.0005 de cut = 0.0889+/-0.0004 uds : mES cut = 0.1395+/-0.0006 de cut = 0.0961+/-0.0002

  34. BK: selection efficiency K candidate from Bsig |cos*trk|<0.85 NextraTrk<3 Eextra<0.4 GeV N0=0 pKB > 1.1Gev/c -0.85<cospmiss<0.9 • BaBar cut and count analysis • Selection:  = n_selected/n_generated Qtag = ± 1 5.270<mES<5.288 GeV/c2 |cosBreco,Thrust|<0.85 with pK,B and Eextra cuts (DG4-DG3)/DG3 ≈ -19%

  35. BK*: selection efficiency (I) with Eextra cut

More Related